A hybrid named entity recognizer for Turkish with applications to different text genres

Dilek Küçük; Adnan Yazici

doi:10.1007/978-90-481-9794-1_23

A hybrid named entity recognizer for Turkish with applications to different text genres

Dilek Küçük
, Adnan Yazici

Department of Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Citations (Scopus)

Abstract

In this study, we present a hybrid named entity recognizer for Turkish, which is based on a previously proposed rule based recognizer. Since rule based systems for specic domains require their knowledge sources to be manually revised when ported to other domains, we turn the rule based recognizer into a hybrid one so that it learns from annotated data and improves its knowledge sources accordingly. Both the hybrid recognizer and its predecessor are evaluated on the same corpora and the hybrid recognizer achieves comparably better results. The current study is significant since it presents the first hybrid -manually engineered and learning-named entity recognizer for Turkish texts.

Original language	English
Title of host publication	Computer and Information Sciences - Proceedings of the 25th International Symposium on Computer and Information Sciences
Pages	113-116
Number of pages	4
DOIs	https://doi.org/10.1007/978-90-481-9794-1_23
Publication status	Published - 2010
Event	25th International Symposium on Computer and Information Sciences, ISCIS 2010 - London, United Kingdom Duration: Sept 22 2010 → Sept 24 2010

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	62 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	25th International Symposium on Computer and Information Sciences, ISCIS 2010
Country/Territory	United Kingdom
City	London
Period	9/22/10 → 9/24/10

Keywords

information extraction
named entity recognition
Turkish

ASJC Scopus subject areas

Industrial and Manufacturing Engineering

Access to Document

10.1007/978-90-481-9794-1_23

Cite this

A hybrid named entity recognizer for Turkish with applications to different text genres. / Küçük, Dilek; Yazici, Adnan.
Computer and Information Sciences - Proceedings of the 25th International Symposium on Computer and Information Sciences. 2010. p. 113-116 (Lecture Notes in Electrical Engineering; Vol. 62 LNEE).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Küçük, D & Yazici, A 2010, A hybrid named entity recognizer for Turkish with applications to different text genres. in Computer and Information Sciences - Proceedings of the 25th International Symposium on Computer and Information Sciences. Lecture Notes in Electrical Engineering, vol. 62 LNEE, pp. 113-116, 25th International Symposium on Computer and Information Sciences, ISCIS 2010, London, United Kingdom, 9/22/10. https://doi.org/10.1007/978-90-481-9794-1_23

@inproceedings{14a1428cb233481c8cd28c89424b7fd9,

title = "A hybrid named entity recognizer for Turkish with applications to different text genres",

abstract = "In this study, we present a hybrid named entity recognizer for Turkish, which is based on a previously proposed rule based recognizer. Since rule based systems for specic domains require their knowledge sources to be manually revised when ported to other domains, we turn the rule based recognizer into a hybrid one so that it learns from annotated data and improves its knowledge sources accordingly. Both the hybrid recognizer and its predecessor are evaluated on the same corpora and the hybrid recognizer achieves comparably better results. The current study is significant since it presents the first hybrid -manually engineered and learning-named entity recognizer for Turkish texts.",

keywords = "information extraction, named entity recognition, Turkish",

author = "Dilek K{\"u}{\c c}{\"u}k and Adnan Yazici",

year = "2010",

doi = "10.1007/978-90-481-9794-1\_23",

language = "English",

isbn = "9789048197934",

series = "Lecture Notes in Electrical Engineering",

pages = "113--116",

booktitle = "Computer and Information Sciences - Proceedings of the 25th International Symposium on Computer and Information Sciences",

note = "25th International Symposium on Computer and Information Sciences, ISCIS 2010 ; Conference date: 22-09-2010 Through 24-09-2010",

}

TY - GEN

T1 - A hybrid named entity recognizer for Turkish with applications to different text genres

AU - Küçük, Dilek

AU - Yazici, Adnan

PY - 2010

Y1 - 2010

N2 - In this study, we present a hybrid named entity recognizer for Turkish, which is based on a previously proposed rule based recognizer. Since rule based systems for specic domains require their knowledge sources to be manually revised when ported to other domains, we turn the rule based recognizer into a hybrid one so that it learns from annotated data and improves its knowledge sources accordingly. Both the hybrid recognizer and its predecessor are evaluated on the same corpora and the hybrid recognizer achieves comparably better results. The current study is significant since it presents the first hybrid -manually engineered and learning-named entity recognizer for Turkish texts.

AB - In this study, we present a hybrid named entity recognizer for Turkish, which is based on a previously proposed rule based recognizer. Since rule based systems for specic domains require their knowledge sources to be manually revised when ported to other domains, we turn the rule based recognizer into a hybrid one so that it learns from annotated data and improves its knowledge sources accordingly. Both the hybrid recognizer and its predecessor are evaluated on the same corpora and the hybrid recognizer achieves comparably better results. The current study is significant since it presents the first hybrid -manually engineered and learning-named entity recognizer for Turkish texts.

KW - information extraction

KW - named entity recognition

KW - Turkish

UR - https://www.scopus.com/pages/publications/78651577925

UR - https://www.scopus.com/pages/publications/78651577925#tab=citedBy

U2 - 10.1007/978-90-481-9794-1_23

DO - 10.1007/978-90-481-9794-1_23

M3 - Conference contribution

AN - SCOPUS:78651577925

SN - 9789048197934

T3 - Lecture Notes in Electrical Engineering

SP - 113

EP - 116

BT - Computer and Information Sciences - Proceedings of the 25th International Symposium on Computer and Information Sciences

T2 - 25th International Symposium on Computer and Information Sciences, ISCIS 2010

Y2 - 22 September 2010 through 24 September 2010

ER -

A hybrid named entity recognizer for Turkish with applications to different text genres

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this